Abstract Background OS, the “gold standard” endpoint in oncology studies, is challenging to measure in earlier stages of BC because of considerably long survival. Further, during extended patient follow-up, treatment effect on OS can become impacted by multiple factors such as non-malignant causes of death, treatment cross-over, and post-progression treatments, making it difficult to isolate the true treatment effect. As an alternative, surrogate endpoints (SE) such as DFS, EFS, and pCR can be used to ascertain meaningful benefit in true endpoint (TE). While studies have previously explored the association of such SEs with OS in different BC populations, there is a dearth in literature of studies reporting on correlation in the NAdj/Adj settings in the HR+/HER2- BC population. The aim of this study is to conduct a correlation analysis between various SE-TE pairs (DFS/EFS vs OS, pCR vs OS and pCR vs DFS/EFS) in this population, from data reported in studies identified from a systematic literature review (SLR). For each pair of endpoints, both absolute and relative treatment effects were evaluated. Methods An SLR was conducted in NAdj/Adj setting for HR+/HER2- BC to identify studies that reported any two of pCR, DFS/EFS and OS. The search was carried out using MEDLINE, Embase, Cochrane Library databases (2000 onward), four BC-related conference proceedings (2018-2020) and clinical trial registries. Relative treatment effect sizes were quantified in terms of HRs for DFS/EFS and OS, and odds ratios (ORs) for pCR. Absolute treatment effects were analyzed in terms of survival probability of EFS/DFS and OS at landmark time points and pCR rates. A weighted least squares regression analysis was carried out to determine the strength of association between the various TE-SE pairs. The regression analysis was carried out on log scale for relative treatment effects and on linear scale for absolute treatment effects. Pearson correlation coefficient (r) was used to quantify the strength of association; coefficient of determination (R2) was also reported to evaluate goodness-of-fit. Sensitivity analyses were performed by considering linear scale (for relative outcomes only), using inverse of variance as weights, and by removing outliers. Results A total of 135 publications were identified in the SLR. Regression analysis showed that there is a statistically significant positive association (r= 0.91; 95%CI, 0.83-0.96, p <0.0001) between log(HROS) and log (HRDFS/EFS), with R2 = 0.83. The correlation between log (HRDFS/EFS) and log(ORpCR) was not strong (r= 0.24; 95% CI, -0.63-0.84; p, 0.603). Correlation of ORpCR with HROS was not conducted due to inadequate data points (n=3). For the absolute outcomes, landmark DFS/EFS at 1 year and 2 years were each found to be moderately correlated with OS at 4 and 5 years. The correlation between pCR rate and landmark DFS/EFS and OS was not strong. The association between the pairs of the endpoints was consistent across all the sensitivity analyses. Conclusions Results of this regression analysis suggest that relative treatment effect on DFS/EFS is strongly associated with relative treatment effect on OS in HR+HER2- BC and could be used as a surrogate for OS. Correlation analysis for HR(DFS/EFS) vs. HR(OS) and OR(pCR) vs. HR(DFS/EFS)Model# studies# observationsPearson coefficient, r (95% CI)p-valueCoefficient of determination, R2 (95% CI)Slope (95% CI)P-valueHRDFS/EFS vs. HROS24320.91 (0.83-0.96)<0.00010.83 (0.69-0.89)0.99 (0.83-1.16)<.0001ORpCR vs. HRDFS/EFS570.24 (-0.63-0.84)0.60.06 (0-0.47)0.23 (-0.86-1.31)0.6099 Citation Format: Anagha Gogate, Inkyu Kim, Sandip Ranjan, Amit Kumar, Eros Papademetriou, Hitesh Bhandari, Ravi Potluri. Correlation between pathologic complete response (pCR), event-free survival (EFS)/disease-free survival (DFS) and overall survival (OS) in neoadjuvant and/or adjuvant (NAdj/Adj) hormone receptor positive (HR+) and human epidermal growth factor receptor-2 negative (HER2-) breast cancer (BC) [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-12-09.
Read full abstract